The long term goal of the proposed research is to identify viral and cellular factors, unique to human mucosa, which contribute to Epstein-Barr virus (EBV) pathogenesis and to mechanisms of persistence. EBV-associated diseases point to the importance of viral interactions in human mucosa: nasopharyngeal carcinoma and oral hairy leukoplakia. (characterized by EBV latency and an aggressively replicative infection, respectively) are epithelial cell diseases. African Burkitt's lymphoma, with its unusual tissue distribution, is a malignancy of the mucosal-associated lymphoid tissue. Utilizing information gained in the current grant period which suggests for the first time the existence of multiple EBV variants with distinct biologic differences and tissue tropisms, we propose the following aims to reach our long-term goals: 1) Determination of the molecular structure of lytic EBV; 2) Evaluation of the distribution of transforming and lytic variants within epithelial and lymphoid components of the mucosal system; 3) Elucidation of mechanisms for EBV persistence in the epithelial components of the mucosa. Based on the known biologic activity of a spontaneous laboratory mutant deleted for the EBV nuclear antigen 2 (EBNA2) encoding gene, we postulate that similar wild type variants represent the "missing link' in our understanding of EBV pathogenic mechanisms. By molecular cloning and polymerase chain reaction analysis, we will delineate (under aim 1) the genetic structure of EBNA2 deleted virus taken directly from throat washings, thus avoiding the selection bias inherent with passage in lymphocyte culture. Demonstration of deletions and potential gene rearrangements will provide an indication of altered biologic activity. In aim #2, use of select genotypic markers amenable to PCR analysis will allow us to track EBV variants through tissue compartments of the mucosal system, identify tissue tropisms, and locate sites for potential interaction between strains. Finally, we will use EBV strains, isogenic except for their EBNA2 gene, to analyze in experimentally infected epithelium transcription from EBV's putative transforming genes. In infected human epithelial cell implants maintained in nude mice, we will relate patterns of gene expression to the EBNA2 deletion mutant's ability to sustain long-term infection in the absence of a viable B cell component.